1. Strongest Agentic Coding Model

• State-of-the-art on Terminal-Bench 2.0 (82.7%), Expert-SWE (73.1%), and SWE-Bench Pro (58.6%), outperforming GPT-5.4 on complex coding tasks.
• Holds context across large systems, reasons through ambiguous failures, and carries changes through surrounding codebases with fewer tokens.

2. Higher Intelligence at GPT-5.4 Latency

• Co-designed, trained, and served on NVIDIA GB200/GB300 NVL72 systems to match GPT-5.4 per-token latency while performing at a significantly higher level.
• Uses fewer tokens to complete the same tasks, making it more efficient as well as more capable.

3. Powerful for Knowledge Work & Computer Use

• Scores 84.9% on GDPval (44 occupations) and 78.7% on OSWorld-Verified for autonomous computer operation.
• Excels at generating documents, spreadsheets, and reports; naturally moves across finding information, using tools, and checking output.

4. Scientific Research Co-Scientist

• Leading performance on GeneBench, BixBench, and FrontierMath; helped discover a new proof about Ramsey numbers verified in Lean.
• Strong enough to meaningfully accelerate progress at the frontiers of biomedical and mathematical research.

1. Advanced Coding Performance

• Achieves 74.5% on SWE-bench Verified, improving the Claude family's state-of-the-art coding abilities.

• Stronger at:

  • Multi-file code refactoring
  • Large codebase debugging
  • Pinpointing exact corrections without unnecessary edits

• Outperforms Opus 4 and shows gains comparable to jumps seen in past major releases.

2. Improved Agentic & Research Capabilities

• Better at maintaining detail accuracy in long research tasks.
• Enhanced agentic search and step-by-step problem solving.
• Performs reliably across complex multi-turn reasoning tasks.

3. Validated by Real-World Users

• GitHub: Better multi-file refactoring and code adjustments.
• Rakuten Group: High precision debugging with minimal collateral changes.
• Windsurf: One standard deviation improvement on their junior dev benchmark - similar magnitude to Sonnet 3.7 → Sonnet 4.

4. Hybrid-Reasoning Benchmark Improvements

• Improvements across TAU-bench, GPQA Diamond, MMMLU, MMMU, AIME (with extended thinking).
• Stronger robustness in long-context reasoning tasks.